LLM EXPLORER
 LLM Explorer: A Curated Large Language Model Directory and Analytics
Dark Theme
  LLM  List    LLM Hosting    LLM Leaderboards    Blog    Newsfeed    Advertise

Octopus V2 by NexaAIDev

 »  All LLMs  »  NexaAIDev  »  Octopus V2   URL Share it on

  1387   
  Arxiv:2404.01744   Arxiv:2404.11459   Arxiv:2404.19296   Android   Autotrain compatible Base model:finetune:google/gem...   Base model:google/gemma-2b   Conversational   En   Function calling   Gemma   On-device language model   Region:us   Safetensors   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/NexaAIDev/Octopus-v2 

Octopus V2 Benchmarks

LLME Score: 0.21647
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Octopus V2 (NexaAIDev/Octopus-v2)
🌟 Advertise your project 🚀

Octopus V2 Parameters and Internals

Model Type 
open-source language model, function calling, on-device language model
Use Cases 
Areas:
Android system management, Orchestration of multiple devices
Applications:
On-device applications for Android, Function calling across complex scenarios
Primary Use Cases:
Generating individual, nested, and parallel function calls
Additional Notes 
Achieves 99.5% accuracy with an average inference latency of 0.38s, significantly outperforming Microsoft's Phi-3 and Apple's OpenELM models in benchmarks.
Supported Languages 
en (proficient)
Training Details 
Data Sources:
android_functions.txt
Methodology:
Unique functional token strategy
Model Architecture:
Advanced language model tailored for Android API function calling
Release Notes 
Version:
V4
Notes:
Advanced open-source language model with 3B parameters, tailored for MMLU benchmark topics. Enhances query handling and function calling accuracy with functional token design.
Version:
V3
Notes:
Sub-billion parameters, multi-modal capabilities, and multilingual support. Enhanced speed, accuracy, and efficiency.
Version:
V2
Notes:
Engineered for Android devices, introduces functional token strategy, excelling in inference speed and accuracy over RAG methods.
LLM NameOctopus V2
Repository 🤗https://huggingface.co/NexaAIDev/Octopus-v2 
Base Model(s)  Gemma 2B   google/gemma-2b
Model Size2b
Required VRAM5.1 GB
Updated2025-07-21
MaintainerNexaAIDev
Model Typegemma
Model Files  5.0 GB: 1-of-2   0.1 GB: 2-of-2
Supported Languagesen
Model ArchitectureGemmaForCausalLM
Licensecc-by-nc-4.0
Context Length8192
Model Max Length8192
Transformers Version4.39.0.dev0
Tokenizer ClassGemmaTokenizer
Padding Token<pad>
Vocabulary Size256022
Torch Data Typebfloat16

Quantized Models of the Octopus V2

Model
Likes
Downloads
VRAM
Octopus V2 GGUF63501 GB

Best Alternatives to Octopus V2

Best Alternatives
Context / RAM
Downloads
Likes
Gemma 1.1 2B It8K / 5.1 GB250244167
Codegemma 2B8K / 5.1 GB2353986
Gita SastraGPT V18K / 5.1 GB271
Caca Tinny 2B V38K / 5.1 GB41
Pandas Tutor Gemma 2B8K / 5.1 GB91
Gemma Qlora Customer Support8K / 5.1 GB90
... 2B Finetuned Sft Navarasa 2.08K / 10 GB125325
Octopus V28K / 5.1 GB682890
Gemma Ko 1.1 2B It8K / 5.1 GB6481
Bge Reranker V2 Gemma8K / 10 GB2555077
Note: green Score (e.g. "73.2") means that the model is better than NexaAIDev/Octopus-v2.

Rank the Octopus V2 Capabilities

🆘 Have you tried this model? Rate its performance. This feedback would greatly assist ML community in identifying the most suitable model for their needs. Your contribution really does make a difference! 🌟

Instruction Following and Task Automation  
Factuality and Completeness of Knowledge  
Censorship and Alignment  
Data Analysis and Insight Generation  
Text Generation  
Text Summarization and Feature Extraction  
Code Generation  
Multi-Language Support and Translation  

What open-source LLMs or SLMs are you in search of? 51544 in total.

Email us: info@extractum.io. Our Privacy Policy  | Terms and Conditions  | Suggest an improvement.
Our Social Media →  
Original data from HuggingFace, OpenCompass and various public git repos.
Release v20241124
  Support LLM Explorer